Cumulus Labs
About
Cumulus Labs is developing Ion, a C++ inference engine with custom CUDA kernels specifically designed for NVIDIA GH200 hardware. Their focus is on optimizing AI inference performance, achieving high token generation speeds and significantly reducing cold start times.
Technology stack
detected 2026-06-30Emailunknown
Comments
No comments yet. Be the first to share your take.
Frequently asked
What does Cumulus Labs do?
Cumulus Labs is developing Ion, a C++ inference engine with custom CUDA kernels specifically designed for NVIDIA GH200 hardware. Their focus is on optimizing AI inference performance, achieving high token generation speeds and significantly reducing cold start times.
How much funding has Cumulus Labs raised?
Cumulus Labs has raised a total of $63M in funding. The most recent round on record is Series A.
Where is Cumulus Labs headquartered?
Cumulus Labs is headquartered in San Francisco, California, USA.
When was Cumulus Labs founded?
Cumulus Labs was founded in 2022.
What industry does Cumulus Labs operate in?
Cumulus Labs operates in Artificial Intelligence, Cloud Computing, Hardware, AI Hardware, AI Chip, AI Inference Engine.